the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 25 Jun 2026
Temporal variability in offshore Fe fluxes in the Peruvian oxygen minimum zone across an El Niño termination
Abstract. Offshore fluxes of Fe in the Peruvian oxygen minimum zone are sensitive to seawater redox state, and potentially modulated by the El Niño-Southern Oscillation. However, observational data on what controls such temporal variation in offshore Fe fluxes remain scarce. Here, we report seawater concentrations of dissolved Fe (dFe), particulate Fe (pFe), and their isotopic compositions (δ56dFe and δ56pFe) from two transects at 12° S and 14° S across the Peruvian shelf. Our data show clear shelf-to-slope dFe and pFe plumes with negative δ56Fe signatures. Maximum Fe concentrations at 12° S decreased from 13 to 4 nmol kg−1 (dFe) and 122 to 13 nmol kg−1 (pFe) over a month across an El Niño event termination. We link these variations to an intensifying Peru-Chile Undercurrent which enhanced oxygen supply and thereby weakened sedimentary reductive Fe effluxes. Off shelf, elevated δ56dFe and low δ56pFe in anoxic waters are attributed to organic ligand binding and authigenic Fe formation.
- Preprint
(11413 KB) - Metadata XML
-
Supplement
(1031 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2026-3486', Anonymous Referee #1, 24 Aug 2026
-
AC1: 'Reply on RC1', Ruifang Xie, 29 Sep 2026
Dear Reviewer #1, we would like to sincerely thank you for reviewing our manuscript and your constructive comments. In this response letter, referee comments are listed, followed by our responses in bold.
This manuscript presents a valuable dataset of dissolved and particulate Fe concentrations and their isotopic compositions in the Peruvian oxygen minimum zone during a post-El Niño period. Through comparison among multiple cruises conducted by the authors, together with comparisons with previously published cruises, the study highlights how the El Niño-Southern Oscillation influence sedimentary Fe release and the subsequent offshore Fe flux. These observations provide valuable constraints on the temporal variability of Fe cycling in oxygen-deficient marine environments and its sensitivity to climate-driven changes in ocean circulation and redox conditions. The manuscript is generally well organized and clearly written. However, several aspects of the interpretation and proposed mechanisms would benefit from further clarification and discussion.
Major concerns:
(1) In Discussion 4.1, the authors found that dFe and pFe were transported farther offshore in the Peruvian OMZ during their sampling period than during the GP16 cruise, and attributed this difference primarily to variations in upwelling intensity. However, it remains unclear whether the dFe efflux from shelf sediments differed between the two periods. In addition, variations in horizontal current velocity between the two periods could also influence the offshore transport of Fe. Could the authors provide evidence or further discuss whether these factors differed between the two periods?
Response: The reviewer brought up a good point that temporal variation in sedimentary dFe efflux would impact offshore dFe transport. Benthic diffusive Fe2+ fluxes from sediments along the 12°S transect have been determined during both cruises of M136 and M137 in Plass et al. (2020, Biogeosciences), which ranged from 17.45 mmol m-2 yr-1 from the inner shelf (nitrogenous in bottom water) to 0.26 mmol m-2 yr-1 on the slope (suboxic in bottom water). Sedimentary Fe2+ effluxes have not been reported for the GP16 cruise, which was carried out during a neutral ENSO phase in austral spring 2013. Conversely, Scholz et al. (2016) have reported overall higher benthic Fe2+ fluxes (1.51 to 22.69 mmol m-2 yr-1) from the 12°S inner shelf (bottom depths <250 m) during cruise M92 (austral summer 2013) than those during M136 and M137 (0.81 to 17.45 mmol m-2 yr-1). Higher benthic Fe2+ fluxes during M92 have been linked to sulfidic conditions on shelf sediments in the neutral ENSO phase (Schlosser et al., 2018, Scientific Reports). Accordingly, we acknowledge that interannual variabilities in benthic diffusive Fe2+ fluxes exist, yet we cannot directly assess this difference between GP16 and M136/M137 due to lack of benthic Fe2+ flux from the former cruise.
We agree with the reviewer that cross-shelf current velocities would impact offshore transport efficiency of Fe (and potentially other dissolved and particulate constituents). Such an effect has been shown in an earlier modeling study for the Peruvian shelf (see the Figure below, Siedlecki et al., 2012, GRL). In this modeling study, weakened upwelling (or downwelling) leads to enhanced surface shoreward currents (cooler colors in subplot b) and benthic offshore current velocities (warmer colors in subplot f), which in turn delivers more sediment-derived dFe offshore (subplot h). These modeling results agree with the upwelling scenarios for GP 16 and M136/M137. Xie et al. (2020) have demonstrated that upwelling rates for M136/M137 were an order of magnitude weaker than for GP16 (Kadko, 2017), potentially allowing more efficient offshore transportation of benthic Fe2+ from reducing sediments.
We have included the above discussion in Section 4.1 in the revised manuscript. Below shows a modeling result from Siedlecki et al. (2012) for the reviewer’s reference.
This figure is from Siedlecki et al. (2012)
References:
Plass, A., Schlosser, C., Sommer, S., Dale, A. W., Achterberg, E. P., & Scholz, F. (2020). The control of hydrogen sulfide on benthic iron and cadmium fluxes in the oxygen minimum zone off Peru. Biogeosciences, 17(13), 3685-3704.
Scholz, F., Löscher, C. R., Fiskal, A., Sommer, S., Hensen, C., Lomnitz, U., ... & Canfield, D. E. (2016). Nitrate-dependent iron oxidation limits iron transport in anoxic ocean regions. Earth and Planetary Science Letters, 454, 272-281.
Schlosser, C., Streu, P., Frank, M., Lavik, G., Croot, P. L., Dengler, M., & Achterberg, E. P. (2018). H2S events in the Peruvian oxygen minimum zone facilitate enhanced dissolved Fe concentrations. Scientific Reports, 8(1), 12642.
Siedlecki, S. A., Mahadevan, A., & Archer, D. E. (2012). Mechanism for export of sediment‐derived iron in an upwelling regime. Geophysical Research Letters, 39(3).
Xie, R. C., Le Moigne, F. A. C., Rapp, I., Lüdke, J., Gasser, B., Dengler, M., Liebetrau, V., and Achterberg, E. P.(2020) Effects of 238U variability and physical transport on water column 234Th downward fluxes in the coastal upwelling system off Peru. Biogeosciences, 17, 4919-4936.
Kadko, D. (2017) Upwelling and primary production during the U.S. GEOTRACES East Pacific Zonal Transect. Global Biogeochem. Cycles, 31, 218-232
In addition, the 14°S section differs from the 12°S section, particularly in terms of the pFe/pAl ratios. The authors attribute this difference mainly to variations in sedimentary reducing dFe efflux between 12°S and 14°S, potentially related to differences in the width of the continental shelf. However, the maximum dFe concentrations along the 14°S section do not appear to be much lower than those at 12°S (e.g., TM94 vs. TM56), which may suggest that the sedimentary dFe efflux at 14°S is not necessarily lower. Could differences in the transformation of dFe to pFe also contribute to the observed contrast in pFe/pAl ratios? In addition, TM94 at 14°S is much closer to the continent. Could an additional lithogenic particulate Fe input contribute to the relatively lower pFe/pAl ratios? Lithogenic proxies, such as particulate Al or Ti, may help constrain this possibility.
Response: Concentrations of dFe in coastal water could be comprised of Fe(II) derived from anoxic shelf sediments, and dFe from subsurface and upwelled water masses. It is Fe(II) concentration that can unmistakably infer sedimentary reducing dFe efflux, which is available only for the 12°S and not the 14°S transect. Upon diffusion from reducing sediments, part of the Fe(II) is transformed to Fe (oxyhydr)oxides, with the rest remaining in dissolved form. In the anoxic shelf waters, this transformation of dFe to pFe is dominated by authigenic pFe mineral formation, resulting in increases in pFe/pAl ratios in particles, e.g., particulate samples at 12°S. The relatively low pFe/pAl ratios at 14°S imply limited authigenic Fe formation. Accordingly, we interpret higher total concentrations of both dFe and authigenic pFe at 12°S to indicate larger Fe(II) fluxes from reducing sediments compared to that at 14°S, the latter of which lacks authigenic pFe signature (i.e., pFe/pAl in proximity to lithogenic endmember value).
Additional terrigenous input is another potential explanation for the low pFe/pAl ratios at 14°S. Indeed, pAl concentrations at 14°S were overall an order of magnitude higher than those at 12°S, which could potential mask authigenic pFe signatures. We have thus modified relevant discussion in the revised manuscript to include this possibility.
Another possibility is that the relatively high pFe/pAl ratios along the 12°S section may partly reflect a biogenic Fe contribution. A pronounced secondary fluorescence maximum occurs near the oxic–anoxic interface, particularly at TM57 and TM56, where relatively high pFe/pAl ratios are also observed. Ohnemus et al. (2016) reported elevated trace metal concentrations in the ODZ during the GP16 cruise and suggested the presence of low-light, autotrophic communities. Biogenic proxies, such as particulate P and C, may provide useful constraints on this possibility.
Response: We appreciate the reviewer’s suggestion to evaluate the contribution of biogenic Fe to pFe/pAl at 12°S. We now estimated concentrations of biogenic Fe in the anoxic waters at TM56 and TM57, the stations that display the highest subsurface pFe/pAl values. At TM56 and TM57, particulate P concentrations at depths of 50-200 m where pFe/pAl ratios were elevated, averaged 46 ± 20 nM (mean ± one SD, n=6). Taking an average pFe/pP ratio of 52.5:1000 in prokaryotes (Cunningham and John, 2017, L&O), we estimated biogenic pFe concentrations at 2.4 ± 1.05 nmol kg-1, accounting for ~4% pFe in particulate samples. This estimation demonstrates that biogenic Fe makes minor contribution to pFe in the Peruvian shelf anoxic waters, consistent with the conclusion in Ohnemus et al. (2017, L&O). We have included this related discussion in the revised manuscript.
Reference:
Cunningham, B. R., & John, S. G. (2017). The effect of iron limitation on cyanobacteria major nutrient and trace element stoichiometry. Limnology and Oceanography, 62(2), 846-858.
Ohnemus, D. C., Rauschenberg, S., Cutter, G. A., Fitzsimmons, J. N., Sherrell, R. M., & Twining, B. S. (2017). Elevated trace metal content of prokaryotic communities associated with marine oxygen deficient zones. Limnology and Oceanography, 62(1), 3-25.
(2) In Discussion 4.3, the authors suggest that the relatively heavy δ⁵⁶dFe and lighter δ⁵⁶pFe observed in the anoxic waters at TM66 and TM75 are related to ligand effects. Is there any additional evidence that the ligands in this depth range have distinctive characteristics that could explain their apparent influence on Fe isotope fractionation?
Response: The reviewer raised a good point. Unfortunately, organic ligand samples were not collected during cruises M136, M137 and M138. Further south of our study site, along a transect at 30°S in the South Pacific Ocean extending from Chile to New Caledonia, equilibrium dFe and labile pFe (LpFe) partitioning has been demonstrated to be a critical pathway for dFe cycling, and that the heterogeneity of organic ligand binding site affinities strongly influence seawater dFe concentrations and potential authigenic particulate Fe formation (Gledhill et al., 2026). The distributions of δ⁵⁶dFe and δ⁵⁶pFe in the anoxic waters at the eddy stations in this study aligned well with those predicted based on field observations both in the Pacific and the North Atlantic (Conway and John et al., 2014; Gledhill et al., 2026; Tagliabue et al., 2023) and laboratory experiments (Morgan et al., 2010). We have modified section 4.3 accordingly to include this discussion.
References:
Gledhill, M., Gosnell, K., Humphreys, M. P., Delaigue, L., Helle, N., Zhu, K., ... & Achterberg, E. P. (2026). Chemical controls on iron distributions across the subsurface South Pacific Ocean. Nature Communications.
Morgan, J. L., Wasylenki, L. E., Nuester, J., & Anbar, A. D. (2010). Fe isotope fractionation during equilibration of Fe− organic complexes. Environmental Science & Technology, 44(16), 6095-6101.
Tagliabue, A., Buck, K. N., Sofen, L. E., Twining, B. S., Aumont, O., Boyd, P. W., ... & Sedwick, P. (2023). Authigenic mineral phases as a driver of the upper-ocean iron cycle. Nature, 620(7972), 104-109.
Conway, T. M., & John, S. G. (2014). Quantification of dissolved iron sources to the North Atlantic Ocean. Nature, 511(7508), 212-215.
Furthermore, the authors discuss the potential influence of reductive and/or non-reductive dissolution of Peruvian slope sediments on deep water Fe. However, the dFe concentrations do not show an obvious increase in the deep waters, which raises questions about the extent to which Peruvian slope sediments influence the deep-water dFe pool. In addition, the authors state that “the δ⁵⁶dFe of −0.5‰ and 0‰ off Peru indicates both non-reductive and reductive release is important…” Could the authors clarify the sources of these two values? Is the −0.5‰ value the minimum δ⁵⁶dFe observed at TM75?
Response: The reviewer is correct that deep waters at offshore stations do not show an obvious increase in dFe compared to nearshore stations; instead, both offshore and nearshore stations have similar dFe concentrations and identical δ⁵⁶dFe within errors (from nearshore to offshore, -0.18 ± 0.52 ‰ at TM56, -0.25 ± 0.11 ‰ at TM66, -0.30 ± 0.40 ‰ at TM75, and -0.14 ± 0.22 ‰ at TM58). These similar dFe concentrations and isotopic signatures indicate they are influenced by similar water masses. Deep water δ⁵⁶dFe values in this study fall within the range between the characteristic δ⁵⁶dFe values for reductive dissolution of -1.2‰ (Chever et al., 2015) and non-reductive dissolution of 0.1‰ (Homoky et al., 2021). In addition, observations from the GP16 transect have shown evident deep dFe, pFe and lithogenic particle plumes extending from the Peruvian slope sediments to offshore waters. Based on this collective evidence, we infer contributions from both reductive and non-reductive dissolution of Fe to deep waters at our offshore stations. We have made corresponding modifications in the revised manuscript.
References:
Chever, F., Rouxel, O. J., Croot, P. L., Ponzevera, E., Wuttig, K., & Auro, M. (2015). Total dissolvable and dissolved iron isotopes in the water column of the Peru upwelling regime. Geochimica et Cosmochimica Acta, 162, 66-82.
Homoky, W. B., Conway, T. M., John, S. G., König, D., Deng, F., Tagliabue, A., & Mills, R. A. (2021). Iron colloids dominate sedimentary supply to the ocean interior. Proceedings of the National Academy of Sciences, 118(13), e2016078118.
Minor comments:
Line 84: The original sentence does not explicitly indicate that M136, M137, and M138 are cruise names. We suggest adding this clarification, e.g., “…showing sampling stations from cruises M136, M137, and M138.”
Response: Agreed and fixed.
Line 103: This sentence mentions “Class 5,” but it is unclear which standard or classification system it refers to. I suggest briefly specifying the relevant standard in parentheses after “Class 5” to improve clarity.
Response: Fixed.
Line 123: Please clarify the source of the Fe(II) data and whether they were collected during the same period and at the same stations as those in this study.
Response: We have added a sentence in the revised manuscript “These Fe(Ⅱ) samples were collected and Fe(II) concentrations were measured on board cruise M136, and data covered stations TM56, TM58 and TM66 (Zhu et al., 2021).” (line 132 in the revised manuscript).
Line 131: “…, while pFe varied from 1 to 17 nmol kg-1 (Figs. 2c and 3)”. The reported increase in pFe does not appear to be clearly supported by Figs. 2c and 3.
Response: We realized this description was misleading, and have modified it as “…, while pFe varied from 1 to 9 nmol kg−1 (near bottom samples not included, Figs. 2c and 3)” (line 139 in the revised manuscript).
Line 140: Since the description of the particulate Fe data includes depth profiles at stations TM77, TM56, and TM75, as well as a few scattered data points at stations TM57, TM66, and TM58, the latter information should also be clarified in the first sentence.
Response: Agreed. We have added a sentence in the revised manuscript “In addition, scatter δ56pFe samples for the surface and near-bottom waters were also collected at stations TM57, TM66 and TM58 (Fig. 2a, c and d).” (line 149 in the revised manuscript)
Line 142: The full name of UCC should be provided when it first appears in the text.
Response: Agreed and fixed.
Lines 150-155: (1) In Fig.3, the location and extent of the continental shelf in the last panel appear inconsistent with those in the first three panels, but all four panels share the same latitude values on the x-axis. Although the authors state that the location of TM56 in panels (a)–(c) is offset for clarity, it would be preferable to use the same transect/location consistently across all four panels. This would avoid potential confusion or misunderstanding when comparing the panels. (2) Please provide more explicit information. For example, the DO contours are shown with grey lines, this should be stated clearly in the figure caption. In addition, please clarify which cruise TM54 and TM60 belong to.
Response: Agreed and fixed.
Line 175: A period (“.”) should be used instead of a comma (“,”) at the end of this sentence.
Response: Corrected.
Lines 182-184: This sentence associates a δ⁵⁶pFe value of approximately −0.2‰ with the high pFe/pAl ratio at TM57. However, there seems no corresponding particulate Fe isotope data available for TM57. Could you please clarify the source of the −0.2‰ value? If direct data from TM57 are unavailable, perhaps the particulate Fe isotope data from TM56 could be used as a reference, given their similar characteristics.
Response: We thank the reviewer for pointing out this typo. Here ‘TM57’ should be ‘TM56’.
Citation: https://doi.org/10.5194/egusphere-2026-3486-AC1 - AC5: 'Reply on RC1', Ruifang Xie, 29 Sep 2026
-
AC1: 'Reply on RC1', Ruifang Xie, 29 Sep 2026
-
RC2: 'Comment on egusphere-2026-3486', Anonymous Referee #2, 24 Aug 2026
The manuscript of Chen et al. investigates iron transfert across the slope offshore Peru during spring 2017. Six vertical profiles of dissolved and particulate iron concentrations with their associated isotopic signature δ56 are described and seem of good quality. Six propositions are done along the manuscript. They are presented by order of importance starting with 1) a plume of sediment-derived dFe originally from the inner shelf extends over more than 100 km offshore ; 2) this plume containt also significant amount of pFe produced in the inner shelf; 3) interannual variations compared to previous cruise (GP16) are due to El Nino event captured in 2017; 4) monthly variations between two legs of the cruise are due to Peru-Chile Undercurrent variations 5) in one station, the plume is not visible due to eddy transport 6) a previous proposition from John et al. 2018 to explain a deeper dFe plume over 1000 km is wrong because others studies have demonstrated that non reductive dissolution of sediment are important persistent sources of dFe.
I have major reservations about the interpretations proposed. Propositions 3 and 4 are presented as hypothesis. Proposition 1 and 5 are not demonstrated which is a particular shame since the demonstration of proposition 1 should be the main goal of the manuscript. The demonstration of proposition 2 presents several contradictions or approximation (in particular with the figure 5). Finally the proposition 6 does not appear to be justified on the basis of the results presented in manuscript. I recommand the authors to reconsider all the discussion focusing on the propositions 1 and 2 but with a solid argumentation.
I will further focus only on the main proposition, proposition 1, whose authors do not discuss its validity using expressions as « striking feature » l. 169 ; « were evidently transported » l.192, « was also evident » l.203 instead. However, an important discussion is required to state that 1a) the dFe of the plume is sediment-derived and 1b) it extend over « 100 km » (l.127) « to the open ocean » (l.192).
To estimate the plume extension, I visualized their dataset of dFe (from Zhu et al 2021) and compared it with GP16 dataset that supposedly missed the plume.The occurrence of a original « 100 km » long plume is not evident… station TM 58 and TM66 seems not related with the plume of dFe event during El nino 2017 while a plume above the slope was also detected in GP16. Station TM55 and TM56 does show a original feature, a more pronounced plume of dFe at about 150m depth not directly connected with the bottom sediment, but it does not extend over 100 km. The high interest of the presented dataset is the significant signal of ligth δ56dFe < -1 ‰ (only visible on TM55 and TM56) that deserves to be discuss based on a solid argumentation (note that, surprisingly, no isotope data at high dFe concentration (> 5nM) is available from GP16). I recommand to the author to only focus on these stations to build a strong discussion about the extension and the origin of the plume.
How to build a solid argumentation concerning the plume origin ? First, the proposition have to be sustained clearly by the dataset. If dFe is transported from the sediment you should be able to draw a mixing line (δ56Fe = f(1/dFe)) along the pynocline of the water sediment interface releasing most of the dFe, in the inner shelf. The figure does not need to be perfect, but it will give the opportunity for the reader to get his/her own appreciation. Second, the results not perfectly consistent with the proposition requires alternative explanations. Why no plume is visible before station TM54 (Figure R1) ? Why very light isotope signal often observed above the OMZ if it is only a signature of sediment effluxes ? If light isotope can origin from another source, why this alternative source not significant in TM55 and TM56 ? Why no plume visible in station TM66 ? only the last question gets an answer in the current manuscript but without the required proves (proposition 5). Third, the proposition should also be defended face to alternative common explanations not even discussed in this version of the manuscript : as micronutrient, is the release of dFe during organic matter remineralisation significant in your system ? probably since you in a very productive area, can you estimate the dFe produced during remineralisation ? as a redox sensitive species, why the in situ reduction of Fe oxides (from atmospheric deposition, or offshore surface transport) negligeable ? if reductive dissolution of Fe oxides occurs in the top of the OMZ, the isotopic signature should be as low as the one you measured ? Only such a reworking of the discussion will be able to convince the readers that the plume observed is originally from the sediment.
Citation: https://doi.org/10.5194/egusphere-2026-3486-RC2 -
AC2: 'Reply on RC2', Ruifang Xie, 29 Sep 2026
Dear Reviewer #2, we would like to sincerely thank you for reviewing our manuscript and your constructive comments. In this response letter, referee comments are listed, followed by our responses in bold.
The manuscript of Chen et al. investigates iron transfert across the slope offshore Peru during spring 2017. Six vertical profiles of dissolved and particulate iron concentrations with their associated isotopic signature δ56 are described and seem of good quality. Six propositions are done along the manuscript. They are presented by order of importance starting with 1) a plume of sediment-derived dFe originally from the inner shelf extends over more than 100 km offshore ; 2) this plume containt also significant amount of pFe produced in the inner shelf; 3) interannual variations compared to previous cruise (GP16) are due to El Nino event captured in 2017; 4) monthly variations between two legs of the cruise are due to Peru-Chile Undercurrent variations 5) in one station, the plume is not visible due to eddy transport 6) a previous proposition from John et al. 2018 to explain a deeper dFe plume over 1000 km is wrong because others studies have demonstrated that non reductive dissolution of sediment are important persistent sources of dFe.
Responses: Please note that in both our original and revised manuscript, we did not state that the reversible scavenging hypothesis by John et al. (2018) was wrong. Instead, we emphasized that our data do not show consistent, strong influence by margin plumes within the anoxic layer, and that these new data are more aligned with the ligand complexation hypothesis.
I have major reservations about the interpretations proposed. Propositions 3 and 4 are presented as hypothesis. Proposition 1 and 5 are not demonstrated which is a particular shame since the demonstration of proposition 1 should be the main goal of the manuscript. The demonstration of proposition 2 presents several contradictions or approximation (in particular with the figure 5). Finally the proposition 6 does not appear to be justified on the basis of the results presented in manuscript. I recommand the authors to reconsider all the discussion focusing on the propositions 1 and 2 but with a solid argumentation.
Response: While we appreciate the time the reviewer put into reviewing our manuscript, these comments are challenging to address in one text. Some of these propositions are relatively well established (e.g. there are several studies showing ‘1’), whereas others (such as propositions 3 and 4) are inevitably more difficult as they require the integration of multiple approaches and will always be subject to more uncertainties.
We would like to stress that we presented in the original manuscript strong evidence to support the temporal variations in benthic dFe supply. Repeat sampling of the same shelf stations at 12°S showed drastic decrease in bottom water dFe and pFe concentrations (Figure 2 in the original manuscript); these changes were accompanied by an intensification of the Peru-Chile Undercurrent (PCUC) driven by a downward displacement of the Coastal Trapped Waves (CTW), which in turn displaced the oxycline, nutricline and thermocline downward thus relaxing bottom water anoxic conditions (Figure S2 and S3 in the revised manuscript). In response, dissolved oxygen (DO) concentrations in bottom waters over the shelf at 200 m depth increased from below detection limit during cruise M136 to over 5 mmol/kg during cruise M137 (Lüdke et al., 2020). These temporal increases in bottom water DO concentrations allow DO diffusion from bottom water to sediment porewater, which would reduce upward Fe(II) flux from reducing sediments. Bottom water DO and H2S levels have been shown to impact benthic Fe(II) fluxes (e.g., Schlosser et al., 2018; Rapp et al., 2020). This discussion is now included in the revised manuscript. Please also refer to our responses to reviewers #1 and #3.
The Fe(II) plume at 12°S (referred to as proposition 1 by the reviewer) are clearly shown in Figures 2 and 3 in the original manuscript, and in Zhu et al. (2021). Iron isotope ratios along this Fe(II) plume are light (−1.34‰ over the shelf and −0.27‰ at the slope) and that at shelf stations were close to the endmember value of sedimentary reductive dissolution. Similarly, the absence of such an Fe(II) plume influence has also been clearly shown in Figure 2 for the eddy station, which displayed much lower dFe and pFe, and heavier δ56dFe values of +0.05 to +0.73‰.
Regarding the comment on proposition 6, we emphasized in the main text that our data do not show consistent, strong influence by margin plumes within the anoxic layer, and that these new data are more aligned with the ligand complexation hypothesis.
References:
Rapp, I., Schlosser, C., Browning, T. J., Wolf, F., Le Moigne, F. A., Gledhill, M., Achterberg, E. P., and Engel, A.: El Niño‐Driven Oxygenation Impacts Peruvian Shelf Iron Supply to the South Pacific Ocean. Geophys. Res. Lett., 47, e2019GL086631
Schlosser, C., Streu, P., Frank, M., Lavik, G., Croot, P. L., Dengler, M., and Achterberg, E. P.: H2S events in the Peruvian oxygen minimum zone facilitate enhanced dissolved Fe concentrations. Sci. Rep., 8, 12642
Zhu, K., Hopwood, M. J., Groenenberg, J. E., Engel, A., Achterberg, E. P., and Gledhill, M.: Influence of pH and dissolved organic matter on iron speciation and apparent iron solubility in the Peruvian shelf and slope region. Environ. Sci. Technol., 55, 9372-9383, https://doi.org/10.1021/acs.est.1c02477, 2021.
I will further focus only on the main proposition, proposition 1, whose authors do not discuss its validity using expressions as « striking feature » l. 169 ; « were evidently transported » l.192, « was also evident » l.203 instead. However, an important discussion is required to state that 1a) the dFe of the plume is sediment-derived and 1b) it extend over « 100 km » (l.127) « to the open ocean » (l.192).
Responses: The “striking feature” referred to here highlights the concurrent shelf-slope plumes of both dissolved (with low δ56dFe) and particulate Fe (with low δ56pFe but higher than δ56dFe), which we interpret to reflect processes that transport both phases of Fe off shelf. In the same paragraph, we discussed that δ56dFe in the elevated Fe(Ⅱ) and dFe plumes and in proximity to shelf sediments have values as low as −1.34‰ that is consistent with reductive dissolution of Fe from Peruvian shelf sediments, thus clearly stating that the dFe in the plume is sediment-derived (lines 170 – 175 in the original manuscript). In the revised manuscript, we further discussed that “benthic diffusive Fe(II) fluxes along the 12°S transect during both cruises of M136 and M137 (Plass et al., 2020) have been reported to range from 17.45 mmol m-2 yr-1 from the inner shelf (nitrogenous in bottom water) to 0.26 mmol m-2 yr-1 on the slope (suboxic in bottom water)” (lines 180-183), thus citing direct evidence from an earlier study conducted during the same campaign to support our inference.
We showed in the original manuscript that such a sediment-derived Fe extended to over 100 km (lines 127-129 and Figure 2 in the original manuscript). Note that this long-distance transport of Fe(II) and dFe has been demonstrated in an earlier study (Zhu et al., 2021). Taking a step forward, we further showed that a strong plume of pFe, sourced from Peru sediments, has also been observed along the 12°S (lines 176-188 in the revised manuscript).
To estimate the plume extension, I visualized their dataset of dFe (from Zhu et al 2021) and compared it with GP16 dataset that supposedly missed the plume.The occurrence of a original « 100 km » long plume is not evident… station TM 58 and TM66 seems not related with the plume of dFe event during El nino 2017 while a plume above the slope was also detected in GP16. Station TM55 and TM56 does show a original feature, a more pronounced plume of dFe at about 150m depth not directly connected with the bottom sediment, but it does not extend over 100 km. The high interest of the presented dataset is the significant signal of ligth δ56dFe < -1 ‰ (only visible on TM55 and TM56) that deserves to be discuss based on a solid argumentation (note that, surprisingly, no isotope data at high dFe concentration (> 5nM) is available from GP16). I recommand to the author to only focus on these stations to build a strong discussion about the extension and the origin of the plume.
Response: We cannot verify the dFe dataset the reviewer downloaded, nor the figure generated by the reviewer, as these are not available in the their report. Please note, the number of dFe profiles obtained over M135 to M138 (back to back cruises from 12°S to 17°S in the Peruvian upwelling zong) are much more extensive than the number of profiles with isotopic data, or Fe(II). The M135 data (17°S) extends furthest off shelf and does show dFe enrichment extending off shelf as stated (Hopwood et al. 2021). This is also evident in the specific stations which we refer to herein that contain additional Fe(II) and isotopic data at 12°S.
We would like to further clarify that in both our original and revised manuscripts, we never stated that GP16 missed the dFe plume. Instead, we stated that “Along the GEOTRACES GP16 section, which was sampled under close to neutral ENSO conditions in austral Spring, 2013, dFe and pFe were shown to be restricted to the Peruvian inner shelf” (lines 213-215). Fe(II) at 12°S off Peru was reported in Zhu et al. (2021) and the original data was documented in Pangaea under Hopwood et al. (2021) (the same data is present in the latest GEOTRACES IDP). We plot below a cross-section plot from Zhu et al. (2021; left panel) and that regenerated in this study using the same dataset (right panel), both showing clear elevated dFe and Fe(II) extending from the Peruvian shelf westward for over 100 km. In contrast to the reviewer’s comment that station TM58 was not influenced by the dFe plume, these figures below indicate that both dFe and Fe(II) at TM58 at ~100 to 350 m (plume depth) were significantly higher than samples shallower and deeper. Conversely, we’ve demonstrated in the original manuscript that eddy station TM66 did not align with the 12°S transect, and that biogeochemistry within this mesoscale eddy was distinctly different from those outside of the eddy.
Reference:
Hopwood, Mark James; Meyer, Judith; Achterberg, Eric Pieter; Gledhill, Martha; Mutzberg, André; Nehir, Münevver (2021): Trace metal data from water samples during METEOR cruise M136 [dataset]. PANGAEA, https://doi.org/10.1594/PANGAEA.928118
Hopwood, Mark James; Meyer, Judith; Achterberg, Eric Pieter; Gledhill, Martha; Mutzberg, André; Nehir, Münevver (2021): Trace metal data from water samples during METEOR cruise M135 [dataset]. PANGAEA, https://doi.org/10.1594/PANGAEA.928117
How to build a solid argumentation concerning the plume origin ? First, the proposition have to be sustained clearly by the dataset. If dFe is transported from the sediment you should be able to draw a mixing line (δ56Fe = f(1/dFe)) along the pynocline of the water sediment interface releasing most of the dFe, in the inner shelf. The figure does not need to be perfect, but it will give the opportunity for the reader to get his/her own appreciation.
Response: The reviewer raises a good point and a testable hypothesis. We show in a figure below that, for samples within the Fe(II) plume, δ56Fe displays a significant linear correlation with 1/dFe (p < 0.05, R2 = 0.62). This good linear correlation supports our inference that Fe(II) and dFe from shelf sediments are effectively transported offshore. We have included this new figure in the revised manuscript.
In the revised Figure 3, we further showed that this Fe(II) plume was transported offshore along isopycnals 25.8 to 26.6 kg/m3.
Second, the results not perfectly consistent with the proposition requires alternative explanations. Why no plume is visible before station TM54 (Figure R1) ? Why very light isotope signal often observed above the OMZ if it is only a signature of sediment effluxes ? If light isotope can origin from another source, why this alternative source not significant in TM55 and TM56 ?
Response: We do not follow this comment from the reviewer. From inner shelf to outer shelf, stations TM60, TM54, TM56 and TM58 all showed significantly elevated concentrations of Fe(II) within the core OMZ (please see Figures 2 and 3 in the original manuscript). Note the no Fe(II) measurement was conducted at station TM57.
Iron released from Peruvian reductive sediments have been shown to be isotopically light with d56dFe ranging from -0.5 to -1.2‰ (Chever et al., 2015; Scholz et al., 2014; John et al., 2018). The d56dFe measured within the Fe(II) plume in this study, with δ56dFe increased from −1.34‰ over the shelf (TM57) to −0.27‰ offshore (TM58), are consistent with earlier findings. Additional potential Fe sources, including suspended sediments, aerosols and non-reductive dissolution of sediments, have d56dFe either higher (e.g., Scholz et al., 2014) or consistent with upper continental values (e.g., Little et al., 2025).
References:
Chever, F., Rouxel, O. J., Croot, P. L., Ponzevera, E., Wuttig, K., and Auro, M.: Total dissolvable and dissolved iron isotopes in the water column of the Peru upwelling regime. Geochim. Cosmochim. Acta, 162, 66-82, https://doi.org/10.1016/j.gca.2015.04.031, 2015.
John, S. G., Helgoe, J., Townsend, E., Weber, T., DeVries, T., Tagliabue, A., Moore, K., Lam, P., Marsay, C. M., and Till, C.: Biogeochemical cycling of Fe and Fe stable isotopes in the Eastern Tropical South Pacific. Mar. Chem., 201, 66-76, https://doi.org/10.1016/j.marchem.2017.06.003, 2018.
Little, S.H., G.F. de Souza, and R.C. Xie, Metal stable isotopes in the marine realm, in Treatise on Geochemistry (Third edition), A. Anbar and D. Weis, Editors. 2025, Elsevier: Oxford. p. 285-332.
Scholz, F., Severmann, S., McManus, J., and Hensen, C.: Beyond the Black Sea paradigm: the sedimentary fingerprint of an open-marine iron shuttle. Geochim. Cosmochim. Acta, 127, 368-380, https://doi.org/10.1016/j.gca.2013.11.041, 2014.
Why no plume visible in station TM66 ? only the last question gets an answer in the current manuscript but without the required proves (proposition 5).
Response: Please note that station TM66 and TM75 were sampled within an anticyclonic mode eddy, where exchange of dissolved and particulate constituents across eddy peripheries is limited. We have explicitly discussed Fe and Fe isotope biogeochemistry at these eddy stations in section 4.3 of the main text, exploring three mechanisms that are in place, namely ligand complexation (highly likely, core OMZ depth), reversible scavenging (unlikely, core OMZ depth), and contributions from both reducing and non-reducing sedimentary release (possible, depth below the core OMZ).
Third, the proposition should also be defended face to alternative common explanations not even discussed in this version of the manuscript : as micronutrient, is the release of dFe during organic matter remineralisation significant in your system ? probably since you in a very productive area, can you estimate the dFe produced during remineralisation ?
Response: We agreed that contribution of dFe associated with organic matter remineralization should be considered in the discussion, yet its relative contribution to dFe is difficult to quantified without POC flux data in and out of the Fe(II) layer. Alternatively, in the revised manuscript, we stated that “… Sinking biogenic pFe formed in the productive surface waters may contribute to these elevated pFe/pAl signals. Assuming particulate phosphorus is overwhelmingly biogenic (Ohnemus et al., 2017) and using a mean Fe/P ratio of 52.5:1000 of prokaryotic cells (Cunningham and John, 2017), the biogenic pFe concentrations at Fe(II) plume depths at shelf stations closest to the margin (TM57 and TM56) are estimated to be 2.4 ± 1.1 nmol kg−1...” (lines 193-201). If all pFe within the Fe(II) plume at shelf stations gets remineralized, this would contribute to 18-50% of dFe observed. These numbers are overly estimated, as sinking particulate organic matter are unlikely fully remineralized over the Peruvian shelf.
Due to lack of POC flux data, we did not include this discussion in the revised manuscript to avoid being speculative.
as a redox sensitive species, why the in situ reduction of Fe oxides (from atmospheric deposition, or offshore surface transport) negligeable ?
Response: There is a misunderstanding here regarding Fe redox chemistry. Just because DO level in seawater is below detection limit does not mean it is ferruginous and favorable for Fe reduction. The redox sequence requires that DO, nitrate and dMn to be completely consumed before microbes would use dFe as an electron receiver. Hence, it is highly unlikely that the oxygen-depleted water over the Peru shelf sediments favors Fe reduction, given the high concentrations of nitrate. This is in contract to porewater geochemistry within Peruvian shelf sediments where DO, nitrate and dMn are depleted in sequence within several to tens of centimeter below the seafloor.
if reductive dissolution of Fe oxides occurs in the top of the OMZ, the isotopic signature should be as low as the one you measured ? Only such a reworking of the discussion will be able to convince the readers that the plume observed is originally from the sediment.
Response: Correct. Biotic dissimilatory Fe(III) reduction has been shown to induce large Fe isotope fractionation, so that porewater-sourced Fe(II) from reductive dissolution of sedimentary is isotopically light. Oxidation of this benthic Fe(II) diffused to overlying water would lead to even lower δ56dFe values, reaching values as low as -1.25‰ reported over the Peruvian shelf or -3.5‰ in the global ocean (e.g., Chever et al., 2015; Severmann et al., 2010).
References:
Chever, F., Rouxel, O. J., Croot, P. L., Ponzevera, E., Wuttig, K., and Auro, M.: Total dissolvable and dissolved iron isotopes in the water column of the Peru upwelling regime. Geochim. Cosmochim. Acta, 162, 66-82, https://doi.org/10.1016/j.gca.2015.04.031, 2015.
Severmann S, McManus J, Berelson WM, and Hammond D.E.: The continental shelf benthic iron flux and its isotope composition. Geochimica et Cosmochimica Acta 74: 3984–4004. https://doi.org/10.1016/j.gca.2010.04.022, 2010
Citation: https://doi.org/10.5194/egusphere-2026-3486-AC2 - AC4: 'Reply on RC2', Ruifang Xie, 29 Sep 2026
-
AC2: 'Reply on RC2', Ruifang Xie, 29 Sep 2026
-
RC3: 'Comment on egusphere-2026-3486', Anonymous Referee #3, 02 Sep 2026
This article aims to improve our understanding of the temporal variability of Fe fluxes in the Humboldt Current System off the Peruvian coast, using a valuable dataset collected around 12°S and 14°S during 2017. The study addresses an important topic and provides potentially valuable insights into the physical and biogeochemical processes controlling Fe distribution and offshore export in the Peruvian oxygen minimum zone. However, I have several concerns that should be addressed to strengthen the interpretation of the results and the main conclusions.
Major revisions:
1. The proposed link between PCUC intensification and reduced benthic Fe(II) efflux is plausible; however, the role of advection needs to be more explicitly considered and distinguished from changes in sedimentary Fe release. Because the PCUC is a poleward-flowing current whose velocity increased substantially between the April and May occupations, sampling at the same repeat stations does not necessarily imply that the same water masses, or water masses with the same biogeochemical history, were sampled during both periods. Enhanced PCUC transport may alter Fe concentrations through changes in water-mass provenance, alongshore transport, and residence time, independently of, or in addition to, changes in sedimentary Fe release. Therefore, the observed decreases in dFe and pFe cannot be attributed solely to reduced reductive dissolution in the sediments based on the present evidence. This issue is particularly relevant because Lüdke et al. (2020) interpreted the contemporaneous decrease in phosphate partly as a consequence of enhanced advection and reduced time for the accumulation of sediment-derived phosphate. Therefore, the decrease in PO₄ may not constitute independent evidence for reduced sedimentary Fe efflux, as suggested here. I recommend that the authors explicitly evaluate the potential contribution of PCUC advection and changes in water-mass properties to the observed Fe variability. In particular, the physical evidence supporting the proposed mechanism (e.g., changes in PCUC velocity, isopycnal displacement, water-mass properties, and near-bottom oxygen conditions between occupations) should be presented or quantitatively related to the Fe observations. If the available data do not allow the advective and benthic contributions to be separated, this limitation should be acknowledged and the conclusion that the PCUC regulates sedimentary Fe(II) efflux should be stated more cautiously.
2. Figure 5 presents the conceptual mechanism proposed to explain temporal changes in the efficiency of the open-marine Fe shuttle. However, the schematic appears to attribute most of the variability to changes in PCUC intensity and associated bottom-water oxygenation, while upwelling and offshore transport are represented as essentially unchanged between the two scenarios. This simplification is difficult to reconcile with other results presented in the manuscript. The authors explicitly identify offshore eddy stations and discuss distinct Fe concentrations and Fe-isotope signatures associated with these mesoscale structures, including possible lateral supply of sediment-derived Fe. Thus, mesoscale circulation does not appear to be merely a hypothetical process in this region, but rather a component of the circulation sampled during the cruises and potentially relevant to the observed Fe distributions. If the manuscript aims to explain changes in the efficiency of the open-marine Fe shuttle, the processes controlling cross-shelf and offshore transport should be considered together with changes in sedimentary Fe release. Mesoscale eddies, filaments, and frontal circulation may influence the trapping, redistribution, and offshore export of sediment-derived Fe, as well as water-mass residence times and oxygen distributions. Therefore, it is unclear whether the substantially larger Fe plume observed in April relative to May reflects only enhanced sedimentary Fe release under more reducing conditions, or also differences in lateral transport and mesoscale circulation between the two occupations. I recommend that the authors clarify the role of the observed eddies and, where possible, evaluate whether mesoscale/cross-shelf circulation differed between April and May. Figure 5 should either incorporate these transport pathways or clearly state that it represents only the hypothesized PCUC–oxygen–sedimentary Fe mechanism rather than the complete physical framework controlling the open-marine Fe shuttle. Similarly, the representation of unchanged upwelling and offshore transport between the two PCUC states should be justified or identified explicitly as a schematic assumption.
Minor revisions:
Line 34: Change “aera” to “area”.
Lines 85–86: Please include the sampling dates of the stations either on the map or in the figure legend. Also indicate the date corresponding to the chlorophyll data shown in the figure.
Lines 93–96: Please rewrite this section to clearly describe the stations sampled during each cruise and provide the corresponding cruise dates. Please also indicate the El Niño conditions prevailing during each sampling period according to the Coastal El Niño Index (ICEN).
Figure 2: Please use the same colors and symbols as in Figure 1b for the profiles shown in Figure 2.
Figures: Please use consistent colors and symbols throughout all figures to facilitate comparison among stations, cruises, and sampling periods.
Figure 3: Why is the inset map in Figure 3a different from the map shown in Figure 1? Please make the maps consistent or clarify the reason for the difference.
Figures 3 and 4: Please indicate when the background oxygen data shown in these figures were collected and whether they correspond to the same sampling periods as the Fe observations.
Citation: https://doi.org/10.5194/egusphere-2026-3486-RC3 -
AC3: 'Reply on RC3', Ruifang Xie, 29 Sep 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2026/egusphere-2026-3486/egusphere-2026-3486-AC3-supplement.pdf
-
AC3: 'Reply on RC3', Ruifang Xie, 29 Sep 2026
Data sets
Dissolved and particulate Fe concentrations and Fe isotope compositions from the Peruvian upwelling zone (M136 to M138) G. Chen et al. https://doi.pangaea.de/10.1594/PANGAEA.995338
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 226 | 116 | 48 | 390 | 39 | 31 | 24 |
- HTML: 226
- PDF: 116
- XML: 48
- Total: 390
- Supplement: 39
- BibTeX: 31
- EndNote: 24
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This manuscript presents a valuable dataset of dissolved and particulate Fe concentrations and their isotopic compositions in the Peruvian oxygen minimum zone during a post-El Niño period. Through comparison among multiple cruises conducted by the authors, together with comparisons with previously published cruises, the study highlights how the El Niño-Southern Oscillation influence sedimentary Fe release and the subsequent offshore Fe flux. These observations provide valuable constraints on the temporal variability of Fe cycling in oxygen-deficient marine environments and its sensitivity to climate-driven changes in ocean circulation and redox conditions. The manuscript is generally well organized and clearly written. However, several aspects of the interpretation and proposed mechanisms would benefit from further clarification and discussion.
Major concerns:
(1) In Discussion 4.1, the authors found that dFe and pFe were transported farther offshore in the Peruvian OMZ during their sampling period than during the GP16 cruise, and attributed this difference primarily to variations in upwelling intensity. However, it remains unclear whether the dFe efflux from shelf sediments differed between the two periods. In addition, variations in horizontal current velocity between the two periods could also influence the offshore transport of Fe. Could the authors provide evidence or further discuss whether these factors differed between the two periods?
In addition, the 14°S section differs from the 12°S section, particularly in terms of the pFe/pAl ratios. The authors attribute this difference mainly to variations in sedimentary reducing dFe efflux between 12°S and 14°S, potentially related to differences in the width of the continental shelf. However, the maximum dFe concentrations along the 14°S section do not appear to be much lower than those at 12°S (e.g., TM94 vs. TM56), which may suggest that the sedimentary dFe efflux at 14°S is not necessarily lower. Could differences in the transformation of dFe to pFe also contribute to the observed contrast in pFe/pAl ratios? In addition, TM94 at 14°S is much closer to the continent. Could an additional lithogenic particulate Fe input contribute to the relatively lower pFe/pAl ratios? Lithogenic proxies, such as particulate Al or Ti, may help constrain this possibility.
Another possibility is that the relatively high pFe/pAl ratios along the 12°S section may partly reflect a biogenic Fe contribution. A pronounced secondary fluorescence maximum occurs near the oxic–anoxic interface, particularly at TM57 and TM56, where relatively high pFe/pAl ratios are also observed. Ohnemus et al. (2016) reported elevated trace metal concentrations in the ODZ during the GP16 cruise and suggested the presence of low-light, autotrophic communities. Biogenic proxies, such as particulate P and C, may provide useful constraints on this possibility.
(2) In Discussion 4.3, the authors suggest that the relatively heavy δ⁵⁶dFe and lighter δ⁵⁶pFe observed in the anoxic waters at TM66 and TM75 are related to ligand effects. Is there any additional evidence that the ligands in this depth range have distinctive characteristics that could explain their apparent influence on Fe isotope fractionation?
Furthermore, the authors discuss the potential influence of reductive and/or non-reductive dissolution of Peruvian slope sediments on deep water Fe. However, the dFe concentrations do not show an obvious increase in the deep waters, which raises questions about the extent to which Peruvian slope sediments influence the deep-water dFe pool. In addition, the authors state that “the δ⁵⁶dFe of −0.5‰ and 0‰ off Peru indicates both non-reductive and reductive release is important…” Could the authors clarify the sources of these two values? Is the −0.5‰ value the minimum δ⁵⁶dFe observed at TM75?
Minor comments:
Line 84: The original sentence does not explicitly indicate that M136, M137, and M138 are cruise names. We suggest adding this clarification, e.g., “…showing sampling stations from cruises M136, M137, and M138.”
Line 103: This sentence mentions “Class 5,” but it is unclear which standard or classification system it refers to. I suggest briefly specifying the relevant standard in parentheses after “Class 5” to improve clarity.
Line 123: Please clarify the source of the Fe(II) data and whether they were collected during the same period and at the same stations as those in this study.
Line 131: “…, while pFe varied from 1 to 17 nmol kg-1 (Figs. 2c and 3)”. The reported increase in pFe does not appear to be clearly supported by Figs. 2c and 3.
Line 140: Since the description of the particulate Fe data includes depth profiles at stations TM77, TM56, and TM75, as well as a few scattered data points at stations TM57, TM66, and TM58, the latter information should also be clarified in the first sentence.
Line 142: The full name of UCC should be provided when it first appears in the text.
Lines 150-155: (1) In Fig.3, the location and extent of the continental shelf in the last panel appear inconsistent with those in the first three panels, but all four panels share the same latitude values on the x-axis. Although the authors state that the location of TM56 in panels (a)–(c) is offset for clarity, it would be preferable to use the same transect/location consistently across all four panels. This would avoid potential confusion or misunderstanding when comparing the panels. (2) Please provide more explicit information. For example, the DO contours are shown with grey lines, this should be stated clearly in the figure caption. In addition, please clarify which cruise TM54 and TM60 belong to.
Line 175: A period (“.”) should be used instead of a comma (“,”) at the end of this sentence.
Lines 182-184: This sentence associates a δ⁵⁶pFe value of approximately −0.2‰ with the high pFe/pAl ratio at TM57. However, there seems no corresponding particulate Fe isotope data available for TM57. Could you please clarify the source of the −0.2‰ value? If direct data from TM57 are unavailable, perhaps the particulate Fe isotope data from TM56 could be used as a reference, given their similar characteristics.